The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
High energy pulses of UV light can be especially problematic for creating damage in the surface of fused-silica optics. Even more concerning, once created, damage sites can increase in size (grow) upon exposure to subsequent UV pulses and render the optic useless. For high power laser applications, such as used for fusion energy research conducted in the National Ignition Facility at Lawrence Livermore National Laboratory, the fused-silica optics are expensive and therefore a long service life is very desirable. Surface damage sites in these optics will typically initiate with a diameter in the range of tens of microns and then will grow exponentially in size upon further UV exposure. The size at which a damage site will render an optic useless depends on the specific application, however, a nominal maximum size is about 1 mm. For the present disclosure, the interest is in treating surface damage sites with diameters that are typically less than about 110 um.
Because of its very strong absorption in silicate glasses, carbon dioxide (CO2) lasers operating at 10.6 μm wavelength have been used successfully to improve the damage threshold or to arrest (mitigate) the growth of laser damage in the surface of fused silica optics. CO2 laser-based mitigation can be very effective in repairing the defects associated with a damage site by removal of the damaged or defective material from the optic's surface through evaporation (i.e., by heating to temperatures greater than about 2500K), or by melting and re-flowing the damaged material at temperatures below the evaporation point (i.e., temperatures below about 2500K).
Over the past three decades, several CO2 laser-based mitigation techniques have been successfully developed for controlling the initiation and growth of surface damage sites in fused-silica. For a mitigation technique (herein referred to as a “mitigation protocol”) to be effective, it is strongly preferable that the technique satisfies three acceptability requirements: 1) prevent the re-initiation and/or growth of a damage site upon subsequent exposure to UV laser pulses; 2) leave behind a low enough level of residual stress that nearby features (i.e. flaws and/or cracks) in the surface will not subsequently induce fracture; and 3) have a final physical shape that will not cause unacceptable downstream intensification when a laser beam passes through the mitigated site. In particular, there could be features of the mitigated site that cause constructive interference that produces intensification sufficient to damage downstream optics, or the exit surface of the mitigated optic, should the mitigated site be on its input surface. Under CO2 laser-heating conditions and at peak temperatures above 2200K, thermo-capillary driven flow can produce a final shape to the mitigated site that can adversely refract and diffract UV laser light passing through it upon subsequent exposure. One of the most common diffracting features in the final shape of the mitigated site is a “rim” surrounding the crater produced by the mitigation procedure. These rims frequently are responsible for producing unacceptably high intensification. Furthermore, any surface structure in the mitigated site that would approximate a positive lens could also produce high intensification. It was realized in developing the subject matter of the present disclosure that meeting requirement 3) was an especially significant challenge in developing a successful mitigation protocol.
Previous to 2005, prior art systems and methods addressed only requirements 1) and 2) described above. Indeed, Brusasco et al., “Methods for mitigating surface damage growth in NIF final optics,” L. W., Proceedings of SPIE 4679, 23, in 2002, teach that a single application of a 5 mm, 10.6 um wavelength CO2 laser beam in the power range 17-35 W applied for 1 second was 100% effective at mitigating growth of UV laser-induced surface damage in a fused silica sample. Further, based on this, Hackel et al. in 2003 patented a “Method for producing damage resistant optics” (U.S. Pat. No. 6,518,539). The Hackel et al. invention offers a general description of a CO2 mitigation protocol as a mitigation process on a fused-silica optic that is performed with a CO2 laser to locally soften the material within, and in the immediate vicinity of, each damage site, to thus anneal out each damage site. The Hackel et al. patent describes using a CO2 laser having its power density and duration controlled to thermally soften fused-silica in a way that minimizes vaporization of material and thermally induced stress in the material. However, Hackel et al. makes no mention or suggestion of controlling downstream intensification by controlling the final surface shape of the CO2 laser-mitigated site.
It was not until 2006 that researchers began to appreciate the unique difficulty presented by the downstream intensification produced by the final surface shape of the mitigated damage site. For instance, with Bass et al. in 2006, the mitigation approach was to remove the entire damage site with several high temperature exposures. This procedure involved three scans of a 200 μm diameter laser spot in an inwardly moving spiral over the damage site. In order to remove (ablate) enough material to completely mitigate these sites, large areas of the surface were heated to temperatures between about 2500K and 3000K for the 150 ms duration of the scan. This resulted in re-deposition of some of the ablated material from the site and thermo-capillary driven flow of, silica along the surface of the mitigation pit. However, this re-deposited material was prone to further laser damage. As a result, techniques were developed to re-melt this re-deposited material at lower temperatures. This was done by using the CO2 laser operated at lower power to eliminate the re-deposited material's potential for damage. However, the thermo-capillary driven flow caused the formation of a “bump” at the bottom of the mitigation pit that would very often produce problematic downstream intensification. It was then found that a second pass with a lower power CO2 beam would smooth this “bump” and in certain cases, but not all, alleviate the problematic downstream intensification.
Guss et al., “Mitigation of growth of laser initiated surface damage in fused silica using a 4.6 um wavelength laser,” Proceedings of SPIE 6403, 64030M-1, 2007, investigated using a frequency-doubled CO2 laser operating at a wavelength of 4.6 um to mitigate damage sites with deep cracks. The motivation for using 4.6 um light was a >25 times longer absorption length in fused-silica at room temperature compared to that at 10.6 um. Guss et al. showed that it was possible to mitigate damage sites and subsurface cracks using 4.6 um wavelength light without significant ablation of the material. The resulting depths of the mitigated sites were much shallower than those produced by the method used by Bass et al. in 2005, and had no “bump” at the pit's bottom. This lack of a “bump” was due to the reflow of the site being dominated by surface tension vs. thermo-capillary flow forces. However, in some fraction of the mitigated sites it was observed that a rim was created that produced unacceptably high downstream intensification.
In 2008, Matthews et al., “Downstream intensification effects associated with CO2 laser mitigation of fused silica,” Proceedings of SPIE 6720, 67200A-1, specifically studied the downstream intensification effect associated with four particular CO2 laser mitigation protocols. These protocols were ablative (>2500K) in nature and left mitigation pits with diameters on the order of 500 um. In all four cases, a rim resulted around the pit with heights on the order of about 3 um. In one case a secondary application of the CO2 laser was used, only along the rim, to modulate (“dimple”) the rim and thus reduce its symmetry. Matthews et al. found that the downstream intensification pattern due to the pits generally had two main components—an on-axis “hotspot” and an off-axis “ring caustic”. Both the “hotspot” and “ring caustic” were quantitatively predicted and experimentally shown to have high enough intensification to initiate downstream damage for relevant fluences (˜8 J/cm2) and downstream distances (<150 mm). For the “dimpled” rim case, however, the predicted number of damage sites was essentially zero and no damage was experimentally observed. This implies that the rim was the main culprit driving the intensification and that the resulting intensification could be brought to non-damaging levels by modifying or, in this case, “dimpling” the rim.